This puzzle is inspired by and has similar rules to the series of puzzles by Sleafar.

Definition. A copycat chess opening is a sequence of moves starting from the conventional starting position, where every move by White (including the last one) is copied by Black identically, resulting in a symmetrical position with respect to the mid-horizontal axis. Every move must be legal, but the last move doesn't have to be checkmate, hence the name opening instead of game.

What is the shortest copycat chess opening in which a rook captures another rook?

  • $\begingroup$ Does it need to be symmetrical up to the rook capture or after the rook capture? $\endgroup$ – Brandon_J Apr 7 '19 at 23:10
  • $\begingroup$ @Brandon_J every move, including the last one... $\endgroup$ – Arnaud Mortier Apr 7 '19 at 23:18

How about seven moves each

1) p-a4 p-a5 2) p-b4 p-b5 3) p(b)xp pxp 4) p-a6 p-a3 5) p-a7 p-a2 6) pxN=R pxN=R 7) RxR RxR

  • 1
    $\begingroup$ Made it viewable here: lichess.org/study/CsNAXWW1 $\endgroup$ – Matsemann Apr 9 '19 at 6:20
  • $\begingroup$ A mix of descriptive and algebraic notation $\endgroup$ – trolley813 Apr 9 '19 at 7:13
  • 1
    $\begingroup$ @trolley813 That's because I am old-school - originally learned P-KR4 etc. $\endgroup$ – Penguino Apr 9 '19 at 20:27
  • $\begingroup$ @Penguino Here descriptive is even better since you have to write each move only once (e.g. 1. P-QR4 2. P-QKt4 3. KtPxP 4. P-R6 5. P-R7 6. PxKt=R 7. RxR, I probably write it wrong, since in my country DN isn't used at all). $\endgroup$ – trolley813 Apr 11 '19 at 5:56

I hope this answer doesn't get sniped...

It did get sniped, but @penguino had an answer that I built off of:

1. a4 a5 2. b4 b5 3. bxa5 bxa4 4. a6 a3 5. a7 a2 6. axb8=R axb1=R 7. Rbxa8 Rxa1


We can't have one of white's rooks that's already on the board take one of black's rooks across a file, since that move would not be copyable. Also, it's impossible for the rooks to capture each other along a rank, since that would require the rooks pass each other at some point, which cannot happen as they will mutually block each other each move they make.

The only other way to have a rook takes rook opening would be pawn promotion, whereby the mirrored pawns promote to a rook and then take the already-there rooks. This quickly leads us to the pawns on the a- and b-files (or g- and h-files), and thus, the solution above.

  • $\begingroup$ Yes you did :) +1 though. $\endgroup$ – Arnaud Mortier Apr 7 '19 at 23:27
  • $\begingroup$ Maybe if I added an explanation... $\endgroup$ – HTM Apr 7 '19 at 23:28
  • 1
    $\begingroup$ Regarding your explanation, it's a tiny bit trickier: why couldn't the capture occur horizontally and hence be possible to copy? (with the 4 rooks involved...) $\endgroup$ – Arnaud Mortier Apr 7 '19 at 23:35
  • $\begingroup$ @ArnaudMortier: for a horizontal rook capture a rook would have to move to the opposite side of the board. This can't be mirrored because the rooks would collide. ie assume no other pieces in the way to move a rook from a1 to a6 is possible but then the opponents rook would need to move from a8 to a3 - impossible due to the rook on a6. $\endgroup$ – Chris Apr 8 '19 at 11:54
  • 1
    $\begingroup$ @Chris True. But PilsNot has edited his answer since then to add this point of explanation. $\endgroup$ – Arnaud Mortier Apr 8 '19 at 11:56

Here’s another way, but on the kingside.

1. g4 g5 2. h4 h5 3. gxh5 gxh4 4. h6 h3 5. h7 h2 6. hxg8=R hxg1=R 7. Rhxg1 Rxg8

And while I’m at it, here are the fastest way to do this with all other pieces, minus the kings of course.


1. a4 a5 2. b4 b5 3. bxa5 bxa4


1. b3 b6 2. Bb2 Bb7 3. g3 g6 4. Bg2 Bg7 5. Bxg7 Bxg2


1. Nf3 Nf6 2. Ne5 Ne4 3. Nc6 Nc3 4. Nxb8 Nxb1


1. b4 b5 2. c4 c5 3. cxb5 cxb4 4. b6 b3 5. b7 b2 6. bxc8=Q bxc1=Q 7. Qdxc1 Qxc8

Now I can say that I have predicted the future! ;¥ ;D

  • $\begingroup$ (+1): Wow, I didn't know there could be variations of this puzzle in the way you have shown! $\endgroup$ – Mr Pie Apr 8 '19 at 1:08
  • 1
    $\begingroup$ Bishops not fastest: 1. b3 b6 2. Ba3 Ba6 3. Bxe7 Bxe2 4. Bxf8 Bxf1. $\endgroup$ – Remellion Apr 8 '19 at 3:08
  • 1
    $\begingroup$ It wasn't my intention to make the same puzzle with other pieces. For me the rooks are by far the most interesting because you have to find the trick, otherwise it's not possible (fastest or not). $\endgroup$ – Arnaud Mortier Apr 8 '19 at 9:44
  • $\begingroup$ @ArnaudMortier Actually, the same thing applies to queens too. $\endgroup$ – trolley813 Apr 9 '19 at 7:15
  • 1
    $\begingroup$ @trolley813 well, not exactly the same thing: White rooks can't leave the lower half of the board while Black rooks can't leave the upper half. Queens, however, can move diagonally and therefore reach the opposite half of the board easily. The only reason they can't capture another Queen is that there are only two of them. If there were four like the rooks, then you wouldn't need a promotion. $\endgroup$ – Arnaud Mortier Apr 10 '19 at 21:39

Could it be

1. h4 h5
2. g4 g5
3. hxg5 hxg4
4. Rxh8...?

  • 1
    $\begingroup$ lol nice answer; beat me to it $\endgroup$ – Brandon_J Apr 7 '19 at 23:16
  • $\begingroup$ No idea if it’s right or not but that’s the first thing I thought of haha @Brandon_J $\endgroup$ – El-Guest Apr 7 '19 at 23:17
  • 1
    $\begingroup$ I think both of you forgot to read the rules: every move, including the last one... :) $\endgroup$ – Arnaud Mortier Apr 7 '19 at 23:17
  • $\begingroup$ @Brandon_J this entire puzzling experience has been very meta LOL $\endgroup$ – El-Guest Apr 7 '19 at 23:17
  • 5
    $\begingroup$ "I'm So Meta, Even This Acronym" - XKCD $\endgroup$ – Brandon_J Apr 7 '19 at 23:19


8 moves

lichess study

1. a4 a5 2. b4 b5 3. axb5 axb4 4. Nc3 Nc6 5. b6 b3 6. b7 b2 7. b8=R b1-R 8. Rxb1 Rxb8

  • $\begingroup$ Think I just sniped you, friend! $\endgroup$ – El-Guest Apr 7 '19 at 23:16
  • $\begingroup$ I think both of you forgot to read the rules: every move, including the last one... :) $\endgroup$ – Arnaud Mortier Apr 7 '19 at 23:18
  • $\begingroup$ Weeeelll, I'm late to the party again @ArnaudMortier $\endgroup$ – Brandon_J Apr 7 '19 at 23:27
  • $\begingroup$ @Brandon_J I hope you had fun anyway :) $\endgroup$ – Arnaud Mortier Apr 7 '19 at 23:28
  • $\begingroup$ I did. Thanks for the puzzle! $\endgroup$ – Brandon_J Apr 7 '19 at 23:29

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.